This program project application represents an integrated and focused multi-disciplinary approach to molecular profiling of colon cancer for improved individualized treatment of this disease. Over 50,000 patients die each year in the United Sates of uncontrolled metastatic colon cancer. Despite extensive morphological and molecular information, it is presently impossible to predict outcome accurately for individual patients, or to calibrate therapy to the molecular state of the tumor. Thus, there is an urgent need to discover reliable molecular markers or targets for the diagnosis, prognosis, and treatment of cancer. For the better part of a century physicians have relied on histological and clinical criteria for developing prognosis. Unfortunately, while these criteria predict average outcomes, tumor variability prevents meaningful individual prediction and decisions. Preliminary expression profiling of solid tumors using SAGE, cDNA, and oligonucleotide arrays has yielded volumes of data, yet variation between tumors has yielded little concordance among the different techniques, nor accurate predictors of outcome. The current program project application is distinguished from previous approaches by: (i) determining the genetic and epigenetic profiles, (ii) determining mRNA profiles on the same human tumor sample, and (iii) using biostatistic, clustering and neural net analysis to guide gene prioritization and outcome analysis. Our strategic approach is to: (i) Develop and validate colon-specific cDNA arrays, our ligase detection reaction/universal DNA array, and our EndoV/DNA ligase-based technologies for high-throughput detection of cancer-associated expression changes, mutations, deletions, insertions, gene amplification, LOH events, and methylation changes in colon cancers. (ii) Discover and characterize new cancer-pathways genes in invasive adenocarcinomas and metastases based on integrated analysis of expression and mutational profiles. (iii) Identify the patterns of molecular alterations that accumulate in the p53, cell cycle, apoptosis, Wnt signaling, RPTK signaling, and TGF Beta signaling pathway genes in individual tumors. (iv) Describe the progressive and sequential alterations in global gene expression and in the expression of specific cancer pathway genes during the evolution of colon neoplasia from atypical crypt focus to metastasis. (v) Develop a molecular classification of colon carcinogenesis, based upon genetic, epigenetic and mRNA expression profiling, and employ this taxonomy in predicting outcome or response to particular modes of therapy.